Back

QuantConnect advantages over TradeStation

Hello All,

I'm new to QuantConnect and algo trading in general. I have some limited experience coding and backtesting strategies in TradingView's Pine Script language, but that is about it. I hear algo trader Kevin Davey talk about TradeStation's EasyLanguage, but I don't know much about it's capabilities. To all you QuantConnect power users, out there, what are some of the reasons  you've chosen QuantConnect over TradeStation? If not familiar with TradeStation specifically, maybe you can provide some comparisons to other platfroms you have used. I will greatly appreciate the insights.

Update Backtest







0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


Hi Garrett Grow !

I used tradestation years ago for a few years, so I might be one of several here who could try to give some experience about both TS and QC, even if, as always, your choice will depend on your own goals! That is just my personal insight...

Tradestation used to provide a built-in platform in which one could automate strategies about 2 decades ago: that was a huge advantage at that time for those who wanted to at time backtest a strategy and automate it (sending real orders automatically!)  for a fraction of the cost that many professionnals would pay (Bloomberg terminal...) However, this main reason to work with Tradestation is no more a real advantage: you can now automate on Quantconnect with good brokers (like IB for almost all asset classes and OANDA for FX for instance) for a fraction of the price of Tradestation (trading fees), with better order management, better automation (virtual servers are better than desktop automation for many reason), and so on.

As for the languages, QC allow its users to use very good native programming languages: C# and Python. They both have advantages and some guys here would be able to give much more expertise about the advantages/disadvantages of the two than I could! The important thing is that these are 2 very good (and not proprietary) programming languages, with so much help available everywhere that you can learn rather easily: after some learning curve (programming is not easy for anyone), you would be able to code about everything. Tradestation has its own proprietary language named "Easylanguage". It was (hence its name) a very simple language, aimed at helping traders to code their ideas, and not coders to backtest and trade. The language was not object-oriented for a long time, and the IDE was very limited. It improved in the last years, but programming in C# or Python is so much more powerful than coding in Easylanguage these days, even with the OOEL (Object oriented Easy Language).

To my mind, compared to TS, it would probably be slightly harder to work on QC and learn coding in Python (probably the easiest between Python and C# even if some people may disagree), but if you really want to progress in algo trading, you would not regret it compared to the rather limited proprietary Tradestation IDE.

The fact that TS gives only proprietary data, allows to trade above all on its own brokerage... seems to be so limited compared to QC which gives many more data (except options on futures!), various data feeds, and connect to various brokers... makes QC better than TS for most people who would want to become algo traders / developers.

Nevertheless, it will depend on your own goals: it is just my understanding (and my experience) of both IDE. If you have other questions, please do not hesitate to ask.

 

1

Thanks for the thorough response, Laurent! You confirmed some of the reasons that I have chosen QuantConnect over TradeStation:

  • Cheaper fees.
  • Greater flexibility and community knowlege of Python over EasyLanguage and even OOEL.
  • Greater processing power and reliability of cloud computing over desktop apps.
One thing that does dissappoint me is that, as far as I know, neither QC nor TS support walk forward testing. It would also be very nice to be able to provide a range of parameter values to backtest in series with results output into a single report. For example, I might want to set a up backtest that would output plots with Sharpe Ratio or Profit Factor on the Y axis and RSI entry points on the X axis. I realize that allowing that capability might overburden servers though.
0

Thanks Garrett, we call that feature optimization (chart X-Y sharpe etc) and it's on our radar. Hopefully, we'll get there early next year. Its too much power for most people though who tend to abuse it to overfit to historical data. 

Walk Forward "Testing"/Optimization you can do in QuantConnect today by just scheduling some optimization work at regular intervals then switching to the new best-fit parameters once the analysis is done. I had an idea about how we could make it easier for you so I'll see what we can do in the next weeks.

0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


I'm glad to hear that these features may be forthcoming, Jared. 

On the optimization topic, I hope that the output will inlcude results for all backtests carried out during the optimization process and not just "optimal" parameter set. In my novice opinion, being able to plot and vizualize trends between algo inputs and profit measures will be crucial for ensuring robust optimization. I am personally not as interested in fancy genetic optimization or similar techniques. I think that simpler statistical methods like x-y plots, x-y-z plots, and histograms are usually better ways of exploring data.   

I am interested in experimenting with walk forward optimization though, since I have heard Kevin Davey speak to it's merits.

 

0

Update Backtest





0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


Loading...

This discussion is closed